| From | Greg Kroah-Hartman <> | Subject | [PATCH 4.4 06/50] kvm: cap halt polling at exactly halt_poll_ns | Date | Mon, 14 Mar 2016 10:50:24 -0700 |
| |
4.4-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Matlack <dmatlack@google.com>
commit 313f636d5c490c9741d3f750dc8da33029edbc6b upstream.
When growing halt-polling, there is no check that the poll time exceeds the limit. It's possible for vcpu->halt_poll_ns grow once past halt_poll_ns, and stay there until a halt which takes longer than vcpu->halt_poll_ns. For example, booting a Linux guest with halt_poll_ns=11000:
... kvm:kvm_halt_poll_ns: vcpu 0: halt_poll_ns 0 (shrink 10000) ... kvm:kvm_halt_poll_ns: vcpu 0: halt_poll_ns 10000 (grow 0) ... kvm:kvm_halt_poll_ns: vcpu 0: halt_poll_ns 20000 (grow 10000)
Signed-off-by: David Matlack <dmatlack@google.com> Fixes: aca6ff29c4063a8d467cdee241e6b3bf7dc4a171 Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
--- virt/kvm/kvm_main.c | 3 +++ 1 file changed, 3 insertions(+)
--- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -1961,6 +1961,9 @@ static void grow_halt_poll_ns(struct kvm else val *= halt_poll_ns_grow; + if (val > halt_poll_ns) + val = halt_poll_ns; + vcpu->halt_poll_ns = val; trace_kvm_halt_poll_ns_grow(vcpu->vcpu_id, val, old); }
|